arXiv:2609.33772v2 Announce Type: replace Abstract: Executable environments are critical for post-training agents on tasks that require tool use and multi-step interaction, but constructing executabl...
SkillGym is an automatic pipeline that generates verifiable environments for training skill-use agents. It crawls internet skills, filters for reproducible workflows, and uses a builder‑reviewer process to create difficulty‑controlled tasks with reference solutions and verifiers. The system builds 6.8k environments, collects 19k successful trajectories, and fine‑tunes LLMs from 2B to 122B parameters, improving performance and skill invocation rates.
GRADUS introduces a method for training large language model agents in policy-constrained tool environments by decoupling task design from solutions. It treats a fixed environment as a latent syllabus, extracting guarded behavioral operators and composing them into progressively harder tasks while using corpus-level feedback to target missing outcome combinations. The approach jointly calibrates tasks and solutions against the environment, turning verification into a recovery signal that preserves difficult examples, leading to significant performance gains on benchmarks such as $ au^2$-Bench and CAR-Bench.
SPADE (Self-Play in Adaptive Synthetic Executable Environments) is a reinforcement‑learning framework where a single large language model acts as both an Environment Designer—creating executable, long‑horizon training environments—and a Reasoning Agent—learning to act within those environments. The framework uses a regret signal based on the difference between rewarded performance with and without privileged hints to guide the Designer toward environments that are challenging yet solvable. Experiments show that, when scaled to 30‑billion‑parameter models, SPADE outperforms fixed‑environment baselines by significant margins across math, science, code, and reasoning benchmarks, and improves tool‑use performance on BFCL‑v4 and ACEBench‑Agent. whyItMatters":"By making environment design a learnable component, SPADE enables continuous self‑improvement and demonstrates that adaptive, self‑generated training environments can substantially boost language‑model performance across diverse tasks."
Interlock is a framework that transforms long‑horizon tasks into typed completion graphs, linking evidence, decisions, and required effects through grounding, coupling, and realization edges. It compiles these graphs into a resettable workspace, a public request, and an evaluation rubric, and uses source‑coverage, public‑input, and agent‑as‑judge reviews to ensure that all mandatory requirements are met. Fine‑tuning Qwen3.6 backbones on Interlock‑generated trajectories improves pass rates on AutomationBench, JobBench, and Harvey LAB, showing that the synthesized experience benefits external task environments.
S3Gym is an interactive benchmark designed to evaluate large language models (LLMs) on their ability to self-improve through self-testing, self-judging, and self-improvement. It separates permissive exploration from strict held-out evaluation across seven text-based games with executable environment verifiers. Experiments show that self-improvement varies by task, with different experience incorporation pathways (direct history, summary memory, or parameter training) yielding mixed results and highlighting the need for agents to transform feedback into executable, transferable policies.
arXiv:2608. 06352v1 Announce Type: new Abstract: Training terminal agents requires executable and verifiable tasks that are not merely solvable, but appropriately challenging for learning.
arXiv:2609.27717v1 Announce Type: new Abstract: Human-written agent skills encode rich workflows for real-world problem solving, but are typically used as external inference-time instructions rather...
arXiv:2604. 01687v3 Announce Type: replace Abstract: Anthropic proposes the concept of skills for LLM agents to tackle multi-step professional tasks that simple tool invocations cannot address.
Agent0 is a fully autonomous framework that enables large language model agents to evolve without external data by using a multi‑step co‑evolution process. It pits a curriculum agent against an executor agent, both derived from the same base LLM, where the curriculum agent creates increasingly challenging tasks and the executor learns to solve them. By integrating external tools into the executor’s workflow, the system creates a self‑reinforcing cycle that continuously generates high‑quality curricula, leading to significant gains in reasoning performance—an 18% improvement on mathematical reasoning and 24% on general reasoning for the Qwen3‑8B‑Base model.
Training terminal agents requires executable and verifiable tasks that are not merely solvable, but appropriately challenging for learning. Executable validation establishes feasibility, yet does not reveal how a task behaves relative to a given solver setting.
DynaWeave is an automated framework that converts static tasks into dynamic scenarios by generating exogenous events and using DAGs to manage dependencies. It activates these events during agent interaction and employs a dynamic verifier that updates only relevant evaluation checks. Using DynaWeave, the authors collected over 2,000 verified trajectories, fine‑tuned Qwen3.5 models, and achieved significant performance gains on Gaia2, CostBench, and ToolMaze, even surpassing Gemini‑3.1‑Pro‑Preview on ToolMaze.